JP3680983B2 - Encoder initial position detector - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an initial position detection apparatus for an encoder suitable for use in a servomotor or the like.
[0002]
[Prior art]
A conventional example of an N-bit (Nbit) encoder is shown in FIG. In the figure, 1 is a slit plate, 2 is an LED (light emitting diode), 3 is a light receiving element group, 4 is a resistor, 5 is an I / V (current / voltage) conversion circuit, 6 is a comparator, and 8 is an M series code. This is a digital processing unit comprising a conversion table 80 for converting to a position signal, a counter 81, a latch 82, an adder 84 and the like.
[0003]
The slit plate 1 repeats light shielding and transmission based on an M-sequence signal that is a binary random number signal that makes one cycle per rotation, and only 2 ^N times (2 ^N ) times per rotation at equal intervals. It has A and B pulse patterns that repeat light shielding and transmission. Light emitted from the LED 2 is received by the light receiving element group 3 through the slit plate 1. The light receiving element group 3 receives light that has passed through the M series pattern of the slit plate 1 and converts it into current, and receives light that has passed through the A and B pulse patterns of the slit plate 1. An A-pulse light-receiving element, a B-pulse light-receiving element (hereinafter, collectively referred to as A and B-pulse light-receiving element 32) that obtains an output having a period that is ¼ cycle shifted from the A-pulse period, and an M-sequence light-receiving element N transistors 34 and the like that amplify the current converted at 31.
[0004]
The current output from the N transistors 34 is converted into a voltage by the N resistors 4, and the current output from the A and B pulse light receiving elements 32 is converted into a voltage by the I / V conversion circuit 5. The signal converted into the voltage is compared with the center voltage E of the amplitudes (P-P) of the A and B pulse signals by the comparator 6 and shaped into a pulse. N (Nbit) M-sequence signals and A and B pulses converted to these voltages are input to the digital processing unit 8 and processed as follows.
[0005]
For example, as shown in FIG. 5, the conversion table 80 is an N × N size table for converting an N-bit M-sequence signal into N-bit position data. The A and B pulse counter 81 is a counter that counts up / down by multiplying the A pulse and B pulse whose phases are shifted by 90 degrees as shown in FIG.
[0006]
Processing in the digital processing unit 8 is performed as follows.
First, immediately after the power is turned on, M series / position conversion is performed using the M series / position conversion table 80. When this is finished, the M-sequence / position conversion end flag is set, the latch 82 holds the output of the table 80 as the initial position detection value, and the A and B pulse counter 81 is cleared (clr). The adder 84 adds the latch output of the table output (also referred to as the M-sequence position detection value) and the A and B pulse counter values, and the addition result becomes the position detection value of the encoder.
[0007]
[Problems to be solved by the invention]
In the configuration as described above, when a high resolution is to be obtained in the slit plate 1 having a limited diameter (N is increased), the shape corresponding to the pattern of the slit plate is within the limited irradiation area of the LED 2. Since the light receiving element group 3 must be arranged, the area per light receiving element is reduced. In particular, since the number of output signals of the M series light receiving element is large, the area must be reduced compared to the A and B pulse light receiving elements, and the output signal is very small. Therefore, it is necessary to amplify using the transistor 34 or the like. Thus, when amplifying using a transistor, the transistor has a capacitance between the base and the emitter, and a delay occurs in the output signal with the resistor 4 for current / voltage conversion.
[0008]
Therefore, when the rotating body of the encoder rotates at a high speed, the M-sequence signal is delayed due to the delay, and there is a risk of erroneous detection. For this reason, conventionally, when power is turned on for M-sequence position detection, there is often a limitation in use such that the rotating body of the encoder is below a certain number of rotations.
Therefore, an object of the present invention is to enable accurate position detection without being limited by the number of rotations of the encoder rotor.
[0009]
[Means for Solving the Problems]
In order to solve the above-mentioned problem, in the invention of claim 1, an M-sequence signal output means for outputting a binary random number signal (hereinafter referred to as an M-sequence signal) that makes one cycle by one rotation of the rotor of the encoder; A pulse signal that outputs an arbitrary number of resolutions per rotation of the encoder rotor and outputs a B pulse signal with a duty of 50% and a B pulse signal that is shifted by 1/4 of one period from the A pulse signal. An output unit, a conversion unit that converts the M-sequence signal into position data, and a counting unit that counts the A pulse signal and the B pulse signal, and performs initial position detection using the conversion unit when the power is turned on; After that, by adding the output from the counting means to the initial position detection value, in the encoder that can always detect the position,
A Z pulse signal output means for outputting a Z pulse signal generated at the same position by one pulse per rotation, and a position detection means for detecting a position from the Z pulse signal are provided.
When the rotating body of the encoder is rotating at a low speed including when it is stopped, the initial position is detected using the M series signal. When the rotating body of the encoder is rotating at a high speed, the Z pulse signal is used. An initial position detection is performed.
[0010]
In the first aspect of the invention, whether the rotating body of the encoder is low speed or high speed is obtained by subtracting the previous value from the current value of the output value (counter value) from the counting means in a certain sampling period. The absolute value can be determined by comparing with a predetermined value (the invention of claim 2), or the amount of change in the output value (counter value) from the counting means and the change in the output from the converting means When the quantity is equal, it is possible to judge that the output from the conversion means is correct (invention of claim 3), or during the period from the previous detection of the Z pulse to the current detection. When the output value from the means is equal to the number of A and B pulses corresponding to one rotation of the rotary body of the encoder, the output from the position detection means for detecting the position using the Z pulse can be determined to be correct.
[0011]
That is, according to the first aspect of the invention, when the rotary body of the encoder such as the slit plate is rotating at a low speed when the power is turned on, the position is detected using the M-sequence signal and the encoder is rotating at a high speed. Then, by detecting the position using the Z pulse, the initial position can be detected at a low speed or a high speed. Further, the position detection using the M series reads the N M series signals instantaneously, so that the position detection time is short. On the other hand, the position detection using the Z pulse is worst in that the rotating body of the encoder makes one rotation. It takes time until. However, in the present invention, the position detection using the M-sequence is performed at low speed and the position detection using the Z pulse is performed at high speed. Therefore, even when the Z pulse position detection is performed, the rotator rotates once. This time is short and the time required for initial position detection can be short.
[0012]
According to the second aspect of the present invention, in the initial position detection, since the high / low determination is made when the change amount of the A and B pulse count values is large / small in a certain sampling period, a new high / low determination means is newly provided. It can be realized without providing.
In the invention of claim 3, the M-sequence position detection value is adopted as correct because the change amount of the M-sequence position detection value in a certain sampling period matches the change amount of the A and B pulse count values. . In order to compare the results obtained by different detection means with signals from different light receiving elements, even if one detection means fails due to noise, the other detection means may be normal. The possibility that the encoder erroneously detects the initial position can be greatly reduced. In addition, if the above comparison does not match, if this operation is retried, the fact that the initial position detection value is not easily established means that one of the detection means is not normal, and this causes an encoder failure. It can be detected.
[0013]
In the invention of claim 4, the A and B pulse count values change by the number of pulses corresponding to one rotation after the Z pulse is detected immediately after the power is turned on until the next Z pulse is detected. The Z pulse position detection value is adopted as the initial position detection value. This is because signals from different light receiving elements and results obtained by different detection means are compared, so even if one detection means fails due to noise or the like, the other detection means is normal. In some cases, the possibility of the encoder misdetecting the initial position can be greatly reduced. As in the third aspect of the invention, if retry is made when the comparison does not match, the fact that the initial position detection value is not easily established means that one of the detection means is not normal, and thus the encoder Can be detected.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a block diagram showing a first embodiment of the present invention.
The slit plate 1 has an N-bit M-sequence signal pattern that makes one cycle per rotation, A and B pulse signal patterns that repeat light shielding and transmission at equal intervals, and light only once at the same position per rotation. It is assumed that a transmitted Z pulse signal pattern is formed. The light emitted from the LED 2 passes through the slit plate 1 and is converted into current by the light receiving element group 3. Therefore, the light receiving element group 3 includes N light receiving element rows 31 and a B pulse light receiving element arranged so as to output a waveform whose phase is shifted by 1/4 cycle from the output signals of the A pulse light receiving element and the A pulse light receiving element. A and B pulse light receiving elements 32, a Z pulse light receiving element 33, N transistors 34 for amplifying the current output from the M series light receiving element 31, and the like.
[0015]
The currents output from the N transistors 34 are converted from currents to voltages by the N resistors 4, respectively. The current output from the Z pulse light receiving element 33 is converted into a voltage by the I / V conversion circuit 5A. The Z pulse signal converted into the voltage is compared with a comparison voltage E1 whose output (pulse width of the Z pulse) is about the pulse width of the A and B pulses by the comparator 6A, and the Z pulse signal as shown in FIG. Waveform shaping to signal. Similarly, the current output from the A and B pulse light receiving elements 32 is converted into a voltage by the I / V conversion circuit 5. The A and B pulse signals converted to voltages are compared with the voltage E by the comparator 6 and shaped into A and B pulse signals as shown in FIG.
[0016]
N M-sequence signals, A, B pulse signals, and Z pulse signals are respectively input to the digital processing unit 8, processed as follows, and processed into position detection values.
Similar to FIG. 4, the M-sequence / position conversion table 80 is a table having a size of N rows × N columns to be converted into N-bit position data with respect to input of N M-sequence signals. As in FIG. 4, 81 is a counter that doubles the A and B pulses as shown in FIG. 6 and counts up / down, and "0" is preset when the power is turned on.
[0017]
Here, when the power is turned on, first, the low speed / high speed determination circuit 83 determines whether the slit plate 1 is stopped or rotating at a low speed or rotating at a high speed. When the speed is low, the determination circuit 83 outputs “1” and the switch (SW) 86 selects the table 80 output. When the speed is high, the determination circuit 83 outputs “0” and the switch (SW) 86. To select “0”, that is, the position detection value (0) by the Z pulse. Next, the data selected by 86 is latched as an initial position detection value at the rising timing of the initial position detection end signal output from the logic circuit 85 by the latch 82. The initial position detection end signal is cleared by the logic circuit 85 when the power is turned on, and is set when the M-sequence / position detection conversion is completed at low speed and when a Z pulse is detected at high speed. After the initial position detection end signal is set, the initial position detection value latched by the latch 82 and the output of the counter 81 cleared at the rising edge of the initial position detection end signal are added by the adder 84, and the position detection is always performed. Will be performed.
[0018]
FIG. 2 is a block diagram showing a second embodiment of the present invention. Since only the low-speed / high-speed determination circuit 83 is different from FIG. 1, only this will be described and the description of the others will be omitted.
As shown in the figure, the low speed / high speed judgment circuit 83 is a buffer 832 that holds the previous value of the output of the counter 81, a subtracter 831 that subtracts the current value from the previous value, and an absolute value calculator 833 that calculates the absolute value of the subtraction result. And a comparator 834 and the like. Note that the input A of the comparator 835 indicates a determination threshold value for low speed or high speed.
[0019]
That is, the output of the counter 81 is subtracted from the value of the buffer 832 by the subtracter 831 and the absolute value is obtained by the absolute value calculator 833. This absolute value is the amount of A and B pulse change between samplings. This change amount is compared with a judgment threshold A for low speed or high speed in the comparator 834. Thus, the low speed / high speed judgment circuit 83 outputs “1” if the speed is low and “0” if the speed is high. Here, the threshold A for determining whether the speed is low or high is a constant determined from the response of the M-sequence signal.
[0020]
FIG. 3 is a block diagram showing a third embodiment of the present invention. An M-sequence position detection value check circuit 87 and a Z pulse position detection value check circuit 88 are added to that shown in FIG. Is done.
First, the M-sequence position detection value check circuit 87 is a buffer 871 that holds the previous value of the M-sequence position detection value, a subtracter 872 that subtracts the current value from the previous value of the M-sequence position detection value, and the previous output of the counter 81. The subtracter 875 subtracts the current value from the previous value of the output of the buffer 874, the counter 81 that holds the value, and the output of the subtracter 875 from the output of the subtractor 872, and the output of the subtractor 873 is “ A determination circuit 876 for determining whether it is 0 "or the other is used.
[0021]
The operation of the M-sequence position detection value check circuit 87 will be described.
The M-sequence position detection value that is the output of the table 80 is subtracted from the output of the buffer 871 that holds the previous value by the subtractor 872. This subtraction result is the amount of change in the M-sequence position detection value during sampling. On the other hand, the output of the A and B pulse counter 81 is subtracted from the output of the buffer 874 holding the previous value by the subtractor 875. The subtraction result is the number of A and B pulses between samplings. The subtracter 873 subtracts the number of A and B pulses during sampling from the amount of change in the M-sequence position detection value during sampling. The determination circuit 876 determines that the subtraction result is “0”, that is, the number of A and B pulses between samplings and the amount of change in the M-sequence position detection value match. Outputs “0”. If this output is “1”, the M-sequence position detection value is checked, and if it is low, the logic circuit 85 sets the initial position detection end flag. If the output is “0”, the initial position detection end flag remains cleared even at a low speed, and the M-sequence position detection value check operation is retried with the next sampling.
[0022]
On the other hand, the Z pulse position detection value check circuit 88 is a latch 881 that latches the output of the counter 81 at the timing of the current Z pulse detection, and a latch 882 that latches the output of the counter 81 at the timing of the previous Z pulse detection. A subtracter 883 for subtracting the current value from the value, a determination circuit 884 for determining whether the subtraction result is 2 N (2 ^N ), and the like. The operation is as follows.
After the power is turned on, the output value of the counter 81 when the Z pulse is first detected is latched in the latch 882, and then the output value of the counter 81 when the Z pulse is detected is latched by the latch 881, and the difference is subtracted. It is determined by a device 883. The determination circuit 884 determines whether or not the difference is 2 ^N corresponding to the number of A and B pulses for one rotation of the slit plate 1, and if the result is “1”, the Z pulse position detection check is OK. In the case of high speed, an initial position detection end flag is set by the logic circuit 85. If the output is “0”, the initial position detection end flag is not set even at high speed, and the check operation of the Z pulse position detection value is retried.
[0023]
【The invention's effect】
According to the first aspect of the present invention, the initial position is detected when the power is turned on using the M-sequence signal at a low speed, and the Z position is used at a high speed, so that the initial position can be detected from the stop time to the high speed. .
According to the second aspect of the present invention, it is possible to make a determination of low speed / high speed using the conventional A and B pulse count values without providing a new means for detecting the rotational speed. According to the invention of claim 3, it is determined that the M-sequence position detection value is correct when the M-sequence position detection value in a certain sampling period matches the amount of change in the A and B pulse count values, and the initial position detection value As a result, the noise resistance of the M-sequence position detection can be improved.
According to the invention of claim 4, the A and B pulse count values change by the number of pulses corresponding to one rotation after the Z pulse is detected immediately after the power is turned on until the next Z pulse is detected. Thus, by determining that the Z pulse position detection value is correct and adopting it as the initial position detection value, the noise resistance of the Z pulse position detection can be improved.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing a first embodiment of the present invention;
FIG. 2 is a block diagram showing a second embodiment of the present invention.
FIG. 3 is a block diagram showing a third embodiment of the present invention.
FIG. 4 is a configuration diagram showing a conventional example.
FIG. 5 is an explanatory diagram of a table example for converting an M-sequence signal into position data.
6 is an operation explanatory diagram of FIGS.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Slit board, 2 ... LED (light emitting diode), 3 ... Light receiving element group, 31, 32, 33 ... Light receiving element, 34 ... Transistor, 4 ... Resistance, 5, 5A ... Current / voltage (I / V) conversion circuit 6, 6A, 834 ... comparator, 8 ... digital processing unit, 81 ... counter, 82, 881, 882 ... latch, 83 ... low speed / high speed judgment circuit, 84 ... adder, 85 ... logic 00806 ... switch, 87 ..., M-sequence position detection value check circuit, 88 ... Z pulse position detection value check circuit, 831, 872, 875, 883 ... subtractor, 832, 871, 874 ... buffer, 833 ... absolute value calculator, 876, 884 ... judgment circuit .

Claims (4)

M-sequence signal output means for outputting a binary random number signal (hereinafter referred to as an M-sequence signal) that makes one cycle per one rotation of the encoder rotator, and an arbitrary number of resolutions output per one rotation of the encoder rotator And a pulse signal output means for outputting an A pulse signal having a duty of 50% and a B pulse signal shifted from the A pulse signal by 1/4 of one cycle, and a conversion for converting the M-sequence signal into position data. Means and a counting means for counting the A pulse signal and the B pulse signal. When the power is turned on, the conversion means is used to detect the initial position, and then the initial position detection value is output from the counting means. In the encoder that can always detect the position by adding,
A Z pulse signal output means for outputting a Z pulse signal generated at the same position by one pulse per rotation, and a position detection means for detecting a position from the Z pulse signal are provided.
When the rotating body of the encoder is rotating at a low speed including when it is stopped, the initial position is detected using the M series signal. When the rotating body of the encoder is rotating at a high speed, the Z pulse signal is used. An initial position detection apparatus for an encoder, characterized by performing initial position detection. Whether the rotary body of the encoder is low speed or high speed is compared with a predetermined value for the absolute value of the difference value obtained by subtracting the previous value from the current value of the output value (counter value) from the counting means in a certain sampling period. The initial position detecting apparatus for an encoder according to claim 1, wherein the initial position detecting apparatus is determined based on the above. 2. The output from the conversion means is judged to be correct when the change amount of the output value (counter value) from the counting means is equal to the change amount of the output from the conversion means. The initial position detection apparatus of the encoder described in 1. When the output value from the counting means is equal to the number of A and B pulses corresponding to one rotation of the rotary body of the encoder between the previous detection of the Z pulse and the current detection, the position is determined using the Z pulse. 2. The initial position detecting device for an encoder according to claim 1, wherein the output from the position detecting means to be detected is determined to be correct.

Images